KR101428030B1 - Video decoding apparatus using motion vector - Google Patents

Abstract

The present invention relates to a method and apparatus for encoding / decoding an image using a motion vector.
According to an embodiment of the present invention, a video decoding method is provided. The video decoding method includes generating a cut motion vector by cutting a motion vector of a reference picture into a predetermined dynamic range, storing the cut motion vector in a buffer, and calculating a motion of a block to be decoded And performing inter-prediction decoding using the motion vector of the current block and the motion vector of the current block.
According to the present invention, it is possible to reduce the size of the memory space required to store the motion vector.

Description

TECHNICAL FIELD [0001] The present invention relates to a video decoding apparatus using a motion vector,

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to image processing, and more particularly, to a coding and decoding method and apparatus using a motion vector.

Recently, as broadcasting system supporting HD (High Definition) resolution has been expanded not only in domestic but also in the world, many users are accustomed to high resolution and high image quality, and accordingly, many organizations are spurring development for next generation video equipment . In addition, with the increasing interest in UHD (Ultra High Definition) that supports more than four times the resolution of HDTV in addition to HDTV, compression technology for higher resolution and higher image quality is required.

An inter prediction technique for predicting a pixel value included in the current picture from a preceding picture and / or a succeeding picture for compression of an image, an intra prediction method for predicting a pixel value using pixel information in a picture, An entropy coding technique may be used in which a short code is assigned to a symbol having a high frequency of occurrence and / or appearance frequency, and a long code is assigned to a symbol having a low appearance frequency.

The present invention provides a method and apparatus for encoding / decoding an image using a cut motion vector.

The present invention provides a method of cutting a motion vector of a reference picture.

The present invention provides a method for transmitting information on motion vectors.

An apparatus for decoding an image according to an embodiment of the present invention includes a reference picture buffer for storing a reference picture and a motion compensation unit for generating a prediction block using a motion vector of the reference picture and the reference picture, The motion vector can be cut to a predetermined range.
The motion vector of the reference picture may be resized (scaled) and then cut into the range.
The motion vector may be cut into a predetermined fixed value range.
The motion vector of the reference picture is stored in a predetermined block unit, and the motion compensation unit may generate the prediction block using the motion vector of the reference picture stored in the preset block unit.
The x component and the y component of the motion vector can be cut into the same fixed value range.
The motion vector may be a motion vector of a block decoded in the inter prediction mode.

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

According to the present invention, an image can be encoded / decoded using a cut motion vector.

According to the present invention, it is possible to reduce the size of the memory space required to store the motion vector.

The present invention can reduce the memory access bandwidth required to fetch data from the memory.

1 is a block diagram showing an example of the structure of an image encoding apparatus.
2 is a block diagram showing an example of a structure of an image decoding apparatus.
3 shows an example of a picture to be coded / decoded and a reference picture.
4 is an example of limiting the dynamic range of a motion vector.
5 to 8 are flowcharts showing a method of storing a motion vector of a reference picture.
9 is an example of quantizing a motion vector.
FIGS. 10 to 13 show examples of fetching motion information from a reference picture.
14 is a flowchart illustrating a method of encoding an image according to an embodiment of the present invention.
15 is a flowchart illustrating a method of decoding an image according to an embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In the following description of the embodiments of the present invention, a detailed description of known functions and configurations incorporated herein will be omitted when it may make the subject matter of the present invention rather unclear.

When an element is described as being "connected" or "connected" to another element, it means that it may be directly connected or connected to another element, but another element may be in the middle . In addition, when describing a specific element as " comprising "in the present invention, it is to be understood that additional elements may be included within the scope of the embodiments or technical ideas of the present invention, it means.

The terms "first "," second ", and the like can be used to describe various elements, but the elements are not limited by these terms. That is, the terms are used only for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, the first component may be referred to as a second component, and likewise, the second component may also be referred to as a first component.

In addition, the elements shown in the embodiments of the present invention are shown separately to indicate that they perform different characteristic functions, but it does not mean that each element can not be implemented as one hardware or software. That is, each component is divided for the convenience of description, and a plurality of components may be combined to operate as one component, or one component may be divided into a plurality of components to operate, The present invention is not limited thereto.

In addition, some of the components may be optional components for enhancing performance, not essential components that perform the essential function of the present invention. The present invention can be implemented in a structure including only essential components except for optional components, and a structure including only essential components is also included in the scope of the present invention.

1 is a block diagram showing an example of the structure of an image encoding apparatus.

1, the image encoding apparatus 100 includes a motion prediction unit 111, a motion compensation unit 112, an intra prediction unit 120, a switch 115, a subtractor 125, a transform unit 130, A quantization unit 140, an entropy encoding unit 150, an inverse quantization unit 160, an inverse transformation unit 170, an adder 175, a filter unit 180, and a reference picture buffer 190.

The image encoding apparatus 100 encodes an input image in an intra prediction mode or an inter prediction mode to output a bitstream. Intra prediction is intra prediction, and inter prediction is inter prediction. The image encoding apparatus 100 transitions between the intra prediction mode and the inter prediction mode through the switch 115. [ The image encoding apparatus 100 generates a prediction block for an input block of the input image, and then encodes a residual between the input block and the prediction block.

In the intra-prediction mode, the intra-prediction unit 120 performs spatial prediction using the pixel values of the already-coded blocks around the current block to generate a prediction block.

In the inter prediction mode, the motion predicting unit 111 finds a motion vector in a reference picture stored in the reference picture buffer 190 by searching for a reference block that best matches the input block. The motion compensation unit 112 performs motion compensation using the motion vector to generate a prediction block. Here, the motion vector is a two-dimensional vector used for inter prediction, and represents an offset between the current block of the encoding / decoding and the reference block.

The subtractor 125 generates a residual block based on the difference between the input block and the prediction block, and the transforming unit 130 transforms the differential block to output a transform coefficient. The quantization unit 140 quantizes the transform coefficients and outputs a quantized coefficient.

The entropy encoding unit 150 performs entropy encoding based on the information obtained in the encoding / quantization process to output a bitstream. Entropy encoding reduces the size of a bit stream for a symbol to be encoded by representing a frequently generated symbol in a small number of bits. Therefore, the compression performance of the image can be expected to be improved through entropy encoding. The entropy encoding unit 150 may use an encoding method such as exponential golomb, context-adaptive variable length coding (CAVLC), and context-adaptive binary arithmetic coding (CABAC) for entropy encoding.

A picture encoded for use as a reference picture for performing inter prediction coding needs to be decoded and stored again. Accordingly, the dequantizer 160 dequantizes the quantized coefficients, and the dequantizer 170 performs an inverse transform on the dequantized coefficients to output a recovered differential block. The adder 175 adds the reconstructed difference block to the prediction block to generate a reconstructed block.

The filter unit 180 is also referred to as an adaptive in-loop filter. At least one of deblocking filtering, SAO (Sample Adaptive Offset) compensation, and ALF (Adaptive Loop Filtering) Is applied. Deblocking filtering means eliminating block distortion occurring at the block boundary, and SAO compensation means adding a proper offset to pixel values to compensate for coding errors. Also, ALF means performing filtering based on a value obtained by comparing a reconstructed image with an original image.

On the other hand, the reference picture buffer 190 stores a restoration block that has passed through the filter unit 180. [

2 is a block diagram showing an example of a structure of an image decoding apparatus.

2, the image decoding apparatus 200 includes an entropy decoding unit 210, an inverse quantization unit 220, an inverse transform unit 230, an intra prediction unit 240, a motion compensation unit 250, an adder 255 A filter unit 260, and a reference picture buffer 270.

The video decoding apparatus 200 decodes a bitstream into an intra-prediction mode or an inter-prediction mode and outputs a reconstructed image. The video decoding apparatus 200 transitions between the intra prediction mode and the inter prediction mode through switching of the switches. The video decoding apparatus 200 generates a prediction block by acquiring a difference block from a bitstream, and then adds the difference block and the prediction block to generate a reconstruction block.

The entropy decoding unit 210 performs entropy decoding based on the probability distribution. The entropy decoding process is the opposite of the above-described entropy encoding process. That is, the entropy decoding unit 210 generates a symbol including a quantized coefficient from a bit stream representing a frequently generated symbol in a small number of bits.

The inverse quantization unit 220 inversely quantizes the quantized coefficients, and the inverse transformation unit 230 inversely transforms the inversely quantized coefficients to generate a difference block.

In the intra-prediction mode, the intra-prediction unit 240 performs spatial prediction using the pixel values of the already decoded blocks around the current block to generate a prediction block.

In the inter-prediction mode, the motion compensation unit 250 performs motion compensation using a motion vector and a reference picture stored in the reference picture buffer 270 to generate a prediction block.

The adder 255 adds the prediction block to the difference block, and the filter unit 260 applies at least one of deblocking filtering, SAO compensation, and ALF to the block through the adder to output the reconstructed image.

On the other hand, the reconstructed image is stored in the reference picture buffer 270 and can be used for motion compensation.

Hereinafter, a block means a unit of encoding / decoding. In the encoding / decoding process, the image is divided into a predetermined size and then encoded / decoded. Thus, a block may also be referred to as a coding unit (CU), a prediction unit (PU), a transform unit (TU), or the like, and one block may be divided into sub- It is possible.

Here, the prediction unit means a basic unit of performing prediction and / or motion compensation. The prediction unit can be divided into a plurality of partitions, each of which is called a prediction unit partition. When the prediction unit is divided into a plurality of partitions, the prediction unit partition may be a basic unit of performing prediction and / or motion compensation. Hereinafter, in the embodiment of the present invention, the prediction unit may mean a prediction unit partition.

Meanwhile, HEVC (High Efficiency Video Coding) uses a motion vector prediction method based on Advanced Motion Vector Prediction (AMVP).

In the motion vector prediction method based on the improved motion vector prediction method, not only the motion vector (MV) of the reconstructed block located around the current block to be encoded / decoded but also the motion vector of the current block to be encoded / decoded in the reference picture A motion vector of a block existing in a position or a corresponding position can be used. At this time. A block located at the same position or position corresponding to the block to be hatched / decoded in the reference picture is referred to as a collocated block, a motion vector of the co-located block is referred to as a collocated motion vector, Called a temporal motion vector. However, the collocated block may be a block existing in the same position as the block to be encoded / decoded, that is, a block existing at the corresponding position, have.

In the motion information merge method, motion information is inferred not only from neighboring reconstructed blocks but also from equivalent position blocks, and uses the motion information as motion information of a current block to be encoded / decoded. At this time, the motion information includes inter-prediction mode information indicating a reference picture index, a motion vector, a uni-direction or a bi-direction necessary for inter prediction, a reference picture list ), Prediction mode information about whether the image is coded in the intra-prediction mode or the inter-prediction mode, and the like.

The predicted motion vector predicted in the current block to be coded / decoded includes not only motion vectors of neighboring blocks spatially adjacent to the current block to be coded / decoded, but also motion vectors of neighboring blocks that are temporally adjacent to the current block to be coded / Lt; / RTI >

3 shows an example of a picture to be coded / decoded and a reference picture.

In FIG. 3, a block X represents a block to be encoded / decoded in a picture 310 to be encoded / decoded, and blocks A, B, C, D, and E represent reconstruction blocks . The block T in the reference picture 320 indicates an equivalent position block existing at a position corresponding to the block to be encoded / decoded.

A motion vector predictor index indicates which motion vector is used as a predicted motion vector in a current block to be coded / decoded.

As shown in Table 1, a motion vector predictor index (mvp_idx_l0, mvp_idx_l1) for each reference picture list is transmitted to the decoding apparatus, and the decoding apparatus uses the same motion vector as the predicted motion vector by the encoding apparatus as a predicted motion vector.

When coding / decoding a current block using a motion vector of a neighboring block spatially adjacent to the current block to be coded / decoded, the motion vector can be stored only in a memory having a relatively small size. However, in the case of using a temporal motion vector, since all the motion vectors of the reference picture are required to be stored in the memory, a relatively large memory is required, and the size of the memory access bandwidth required to fetch data from the memory . Therefore, it is necessary to store temporal motion vectors more efficiently in an application environment in which a memory space such as a portable receiving terminal is insufficient or power consumption is to be minimized.

On the other hand, there is a conventional technique of storing a motion vector in a memory to reduce spatial resolution of a motion vector. In the above method, a motion vector is compressed at an arbitrary ratio and stored in a memory. For example, a motion vector stored in units of 4x4 blocks is stored in units of 4x4 or more blocks, and the number of stored motion vectors is reduced. At this time, information on the compression ratio is transmitted in order to adjust the block size of the stored motion vector. The information is transmitted through a Sequence Parameter Set (SPS) as shown in Table 2.

Referring to Table 2, when motion_vector_buffer_comp_flag is 1, a motion vector buffer compression process is performed.

motion_vector_buffer_comp_ratio_log2 represents the compression ratio of the motion vector buffer compression process. If motion_vector_buffer_comp_ratio_log2 does not exist, motion_vector_buffer_comp_ratio_log2 is deduced to be 0, and the motion vector buffer compression ratio is expressed by Equation (1).

For example, when all the 4x4 blocks of the 1920x1080 (1080p) picture have different motion vectors and two reference picture lists are used and two reference pictures are used for each list, a total of 3.21 Mbytes of memory Space is required to store temporal motion vectors.

1. A bit depth of 26 bis for each motion vector,

(1) The dynamic range of the X component of the motion vector: -252 to +7676 (bit depth: 13 bits)

(2) The dynamic range of the Y component of the motion vector: -252 to +4316 (bit depth: 13 bits)

(3) (the dynamic range of each component of the motion vector is calculated based on the first prediction unit in the corresponding picture).

2. When all 4x4 block units have different motion vectors: 480 x 270 = 129600 blocks

3. Use two motion vectors for each block

4. Number of reference picture lists: 2

5. Use two reference pictures per reference picture list

=> 26 bits x 129600 blocks x 2 motion vectors x 2 reference picture lists x 2 reference pictures = 26956800 bits = 3.21 Mbytes

According to the method of reducing the spatial resolution of the motion vector described above, it is possible to reduce the size of the required memory space and the memory access bandwidth using the spatial correlation of the motion vector. However, the above method of lowering the spatial resolution of the motion vector does not limit the dynamic range of the motion vector.

If the size of the memory space is reduced to 1/4, the size of the memory space required in the above example is reduced to about 0.8 Mbytes. At this time, if the dynamic range of the motion vector is additionally restricted and only the bit depth necessary for storing the motion vector is used for each component of the motion vector, the required memory space size can be further reduced to 0.37 Mbytes .

Accordingly, in the present invention, the dynamic range of the motion vector is limited in order to reduce the size of the memory space required to store the motion vector and the memory access bandwidth required to fetch data from the memory. A motion vector of a reference picture with a limited dynamic range can be used as a temporal motion vector in a block to be coded / decoded.

Hereinafter, the dynamic range refers to a period between a minimum value and a minimum value that a negative component or positive component of a motion vector can have based on 0, a bit depth indicates a size of a space required to store a motion vector, May also mean a bit width. Unless otherwise specified, a motion vector means a motion vector of a reference picture, that is, a temporal motion vector.

Each component of the motion vector is expressed as a minimum value or a maximum value of the dynamic range when it deviates from the dynamic range. For example, when the X component of the motion vector is 312 and the maximum value of the dynamic range of each component of the motion vector is 256, the X component of the motion vector is limited to 256.

Likewise, when the bit depth of each component of the motion vector is 16 bits and the motion vector is (-36, 24), by limiting the bit depth of each component of the motion vector to 6 bits, Component has a dynamic range of -32 to +31, and the motion vector is expressed by the dynamic range (-32, 24).

When the bit depth of each component of the motion vector is 16 bits and the motion vector is (-49, 142), if the bit depth of each component of the motion vector is limited to 9 bits, The component has a dynamic range of -256 to +255, and the motion vector is expressed without change (-49, 142).

4 is an example of limiting the dynamic range of a motion vector.

Referring to FIG. 4, if the dynamic range of a motion vector having a dynamic range of -4096 to +4095 is limited to -128 to +127, the bit depth can be reduced from 13 bits to 8 bits.

Each component of the temporal motion vector is clipped as shown in equations (2) and (3) to be stored at a bit depth of N bits (s). Here, N is a positive integer.

Here, MV_X is an X component of a motion vector, MV_Y is a Y component of a motion vector, min (a, b) is an operation of outputting a smaller value among a and b, max (a, b) . clippedMV_X and clippedMV_Y are X and Y components of the temporal motion vector, respectively, and are stored in the memory and used as temporal motion vectors of the current block to be encoded / decoded.

For example, if the size of the memory space is 48 bytes as shown in Table 3, and a bit depth of 16 bits is used for each component of the motion vector, a total of 12 motion vectors can be stored.

MV1-X MV1-Y MV2-X MV2-Y MV3-X MV3-Y MV4-X MV4-Y MV5-X MV5-Y MV6-X MV6-Y MV7-X MV7-Y MV8-X MV8-Y MV9-X MV9-Y MV10-X MV10-Y MV11-X MV11-Y MV12-X MV12-Y

However, if only the bit depth of 8 bits is used for each component of the motion vector, a total of 24 motion vectors can be stored as shown in Table 4.

MV1-X MV1-Y MV2-X MV2-Y MV3-X MV3-Y MV4-X MV4-Y MV5-X MV5-Y MV6-X MV6-Y MV7-X MV7-Y MV8-X MV8-Y MV9-X MV9-Y MV10-X MV10-Y MV11-X MV11-Y MV12-X MV12-Y MV13-X MV13-Y MV14-X MV14-Y MV15-X MV15-Y MV16-X MV16-Y MV17-X MV17-Y MV18-X MV18-Y MV19-X MV19-Y MV20-X MV20-Y MV21-X MV21-Y MV22-X MV22-Y MV23-X MV23-Y MV24-X MV24-Y

Therefore, according to the present invention, an in-loop filtering process such as a deblocking filter or an ad-hoc loop filter can be performed on a video reconstructed by an encoding device and / or a decoding device A motion vector of a reference picture is stored by restricting a dynamic range of a motion vector when the motion vector is stored in a reconstructed picture buffer (DPB). Here, the reconstructed image buffer may mean the reference picture buffer of FIG. 1 or FIG.

I. Motion vector cutting process

The process of cutting each component of the motion vector is performed when the slice type (slice_type) is not an I picture. The motion vector cutting process is performed in units of a tree block or a maximum size coding unit (LCU) after the filtering process.

The input of the motion vector cutting process is (xP, yP), which is the upper left pixel position of the prediction unit in the current picture, and the motion vector matrices MvL0 and MvL1, and the output is the cut motion vector matrices CMvL0 and CMvL1.

The operations of equations (4) to (7) are performed on the matrices MvL0, MvL1, CMvL0 and CMvL1.

Here, TMVBitWidth represents the bit depth of the motion vector, and Clip3 (a, b, c) means a function for cutting c such that it exists within a range between a and b.

II. Motion vector storage process

5 to 8 are flowcharts showing a method of storing a motion vector of a reference picture.

Referring to FIG. 5, a motion vector of a reference picture may be stored using an image buffer storing a reconstructed image and a motion vector buffer storing a motion vector. At this time, the reconstructed image is subjected to an in-loop filtering process (S510), and the dynamic range of the motion vector is limited (S520) and stored (S540).

Referring to FIG. 6, a motion vector is stored (S640) through a dynamic range limitation process (S620) and a spatial resolution reduction process (S630) using an image buffer and a motion vector buffer together.

7, the reconstructed image is stored in an image buffer through an in-loop filtering process S710 (S740), the dynamic range of the motion vector is limited (S720) and stored in a motion vector buffer (S750) do.

8, the reconstructed image is stored in an image buffer through an in-loop filtering process S810, and the motion vector is subjected to a dynamic range limitation process S820 and a spatial resolution reduction process S830 (S850).

6 and 8, the dynamic range limitation process (S620, S820) and the spatial resolution reduction process (S630, S830) are not limited to the order, but may be changed.

In addition, to further reduce the memory access bandwidth, the dynamic range for each component of the motion vector may be differently limited. For example, only one of the dynamic range of the X component and the dynamic range of the Y component can be limited, or the dynamic range of the Y component can be limited more than the dynamic range of the X component.

The limited dynamic range of a motion vector is transmitted via a sequence parameter set, a picture parameter set (PPS) or a slice header, and the decoding apparatus can perform dynamic motion of a temporal motion vector in a sequence, picture or slice The same restriction of the range is performed. At this time, the bit depth, which is the size of the memory space required to store the motion vector expressed in the dynamic range, can be transmitted together. Also, by using a fixed range of bit depths and using a dynamic range transmitted through a sequence parameter, a picture parameter set, or a slice header without storing a motion vector, a temporal motion vector can be efficiently .

On the other hand, the motion vector can be quantized and stored. When the motion vector is quantized and stored, the precision of the motion vector is reduced. The quantization methods include uniform quantization with equal step size, and non-uniform quantization with unequal step size. The step size of the quantization is set to a fixed value preset in the encoding apparatus and the decoding apparatus, or the sequence parameter set. A picture parameter set, a slice header, or the like, from the encoding apparatus to the decoding apparatus. In the decoding apparatus, a quantized motion vector is used as it is or inverse quantized. 9 is an example of quantizing a motion vector. Referring to FIG. 9, if the motion vector has a component value of 32 to 48, the motion vector is quantized to 40.

In addition, the motion vector can be stored with limited display resolution. The rendering resolution means an integer pixel unit (1 pixel unit), a fractional pixel unit (1/2 pixel unit, 1/4 pixel unit, or the like). For example, the resolution of a motion vector processed in units of quarter pixels can be stored as an integer pixel. The representation resolution of the motion vector is set to a fixed value predetermined by the encoding apparatus and the decoding apparatus, or transmitted from the encoding apparatus to the decoding apparatus through a sequence parameter set, a picture parameter set, or a slice header.

In addition, the dynamic range limitation process, the spatial resolution reduction process, and the quantization process of the motion vector can be performed only for some motion vectors among the temporal motion vectors stored in the memory.

When the dynamic range of the motion vector is limited and stored, information on the dynamic range of the motion vector may be added and stored in the memory. For example, when the dynamic range of the motion vector is -128 to +127, a flag of 1 is additionally stored, and when the dynamic range of the motion vector is -32 to +31, a flag of 0 can be additionally stored. At this time, the flag information may be stored together with the motion vector, or may be stored in a memory other than the memory in which the motion vector is stored. When the flag information and the motion vector are stored in different memories, it is possible to arbitrarily access the flag information when finding out what dynamic range the specific motion vector is stored in. In addition, information on which dynamic range a certain motion vector has been stored is transmitted through a sequence parameter set, a picture parameter set, a slice header, or the like, so that the decoder can operate in the same manner as the encoder.

When the spatial resolution of the motion vector is reduced and stored, information on the block size of the motion vector may be added and stored in the memory. For example, when the block size of the motion vector is 4x4, a flag of 1 is additionally stored, and when the block size is 16x16, a flag of 0 can be additionally stored. At this time, the flag information may be stored together with the motion vector, or may be stored in a memory other than the memory in which the motion vector is stored. When the flag information and the motion vector are stored in different memories, it is possible to arbitrarily access the flag information when finding out what block size the specific motion vector is stored in. In addition, information on the block size of some motion vectors is transmitted through a sequence parameter set, a picture parameter set, a slice header, or the like, so that the decoder can operate in the same manner as the encoder.

When the motion vector is quantized and stored, information on the precision of the motion vector can be added and stored in the memory. For example, when the step size of the quantization is 4, a flag of 1 is additionally stored, and when the step size of the quantization is 1, a flag of 0 can be additionally stored. At this time, the flag information may be stored together with the motion vector, or may be stored in a memory other than the memory in which the motion vector is stored. When the flag information and the motion vector are stored in different memories, it is possible to make random access to the flag information when determining which step size the specific motion vector is quantized and stored. In addition, information on how many motion vectors are quantized and stored in which step size is transmitted through a sequence parameter set, a picture parameter set, a slice header, or the like, so that the decoder can operate in the same manner as the encoder.

In addition, when the motion information is stored in the memory, the spatial resolution of the motion vector can be reduced and stored. At this time, the motion information includes inter-prediction mode information indicating a reference picture index, a motion vector, a uni-direction or a bi-direction necessary for inter prediction, a reference picture list ), Prediction mode information about whether the image is coded in the intra-prediction mode or the inter-prediction mode, and the like.

For example, motion information of a prediction unit having a largest partition size among a plurality of motion information of a specific area can be stored in the memory as representative motion information. At this time, the specific area may include an area in a block to be encoded / decoded and a surrounding block area. Also, the specific area may be an area including a block in which motion information is stored, when the picture or the entire slice is divided into a certain size.

For example, representative motion information may be stored in a memory after excluding motion information encoded by a motion information merging method, a skip method, or the like among a plurality of motion information included in a specific area.

For example, motion information that occurs most frequently among a plurality of pieces of motion information included in a specific region may be stored in the memory as representative motion information. At this time, it is possible to calculate the number of occurrences of motion information, etc. according to the size of the block.

For example, motion information of a specific position can be stored in a plurality of pieces of motion information included in a specific area. At this time, the specific position is a position included in the specific region, and may be a fixed position of the specific region. Further, the specific position may be selected as one of the plurality of positions. If a plurality of positions are used, priority can be determined for each position, and motion information can be stored in a memory according to the priority.

For example, when a plurality of pieces of motion information included in a specific area are stored in a memory, motion information is not present outside a block coded in an intra prediction mode, a block coded in a PCM (Pulse Coded Modulation) mode, a slice or a picture boundary Therefore, the motion information at the corresponding position may not be stored in the memory.

If motion information of a specific position is not stored, motion information of an equal position block, motion information of a block to be encoded first, It can be used as information. At this time, the specific position may be one sample position or block position in a block existing in the vicinity of the block to be encoded / decoded. For example, when there is no motion information at a specific position, a median or an average of motion information of inter-prediction-encoded blocks around the corresponding position may be stored in the memory. For example, when there is no motion information at a specific position, an average value of motion information of neighboring blocks at the corresponding position may be stored in the memory. In calculating the intermediate value and the average value, if at least one of the reference picture index, the reference picture list, and the inter prediction mode information is different in motion information of neighboring blocks, the motion vector is a reference picture index. The reference picture list, the inter prediction mode information, the picture order count, and the like.

III. Motion vector derivation process

The motion information may be stored in a memory using the motion information methods described above, and the stored motion information may be obtained when the motion information of the reference picture is used in the motion vector prediction method, the enhanced motion vector prediction method, or the motion information merging method.

For example, motion information at a position corresponding to a position of a block to be encoded / decoded in a reference picture can be obtained. At this time, the position corresponding to the position of the block to be encoded / decoded in the reference picture may be a position relative to the fixed position in the specific area or the position of the block to be encoded / decoded.

FIGS. 10 to 13 show examples of fetching motion information from a reference picture.

Blocks A, B, C, D, and E represent blocks to be coded / decoded in the pictures 1010, 1110, 1210, and 1310 to be coded / decoded. Represents a restoration block located in the periphery of the target block. Blocks T in the reference pictures 1020, 1120, 1220, and 1320 indicate equal-position blocks corresponding to the blocks to be encoded / decoded. Block Y in the reference picture 1320 in Fig. 13 represents a block corresponding to a position other than the current block to be coded / decoded.

Referring to FIG. 10, motion information corresponding to a position corresponding to the upper left pixel position among the positions of the block X to be encoded / decoded in the reference picture may be obtained.

Referring to FIG. 11, motion information corresponding to a position corresponding to a center pixel position among the positions of a block X to be encoded / decoded in a reference picture may be obtained.

Referring to FIG. 12, motion information corresponding to a position corresponding to a lower right pixel position among the positions of a block X to be encoded / decoded in a reference picture may be obtained.

Referring to FIG. 13, motion information corresponding to a position corresponding to a position other than the block X to be encoded / decoded in the reference picture may be obtained.

It is possible to perform a coding / decoding method such as motion vector prediction, enhanced motion vector prediction, motion information merging, and merge skip by using motion information stored in the memory, that is, motion information of a reference picture.

A method of storing a motion vector in a memory using at least one of a dynamic range limitation method of a motion vector, a spatial resolution reduction method of a motion vector, a motion vector quantization method, and a motion vector representation resolution reduction method, It can be used for prediction of a motion vector of a block to be decoded and for merging motion information.

The process of extracting a motion vector of a reference picture from a memory is called a derivation of a temporal motion vector. In the temporal motion vector derivation process, TMVbitWidth represents the bit width of the temporal motion vector stored in the memory.

The input of the temporal motion vector derivation process is (xP, yP) which is the upper left pixel position of the prediction unit in the current picture, nPSW and nPSH which are the horizontal length and vertical length of the luminance prediction unit, and refIdxLX which is the reference picture index of the current prediction unit partition , And the output is the motion vector prediction value mxLXCl and the presence flag availableFlagLXCol.

RefPicOrderCnt (pic, refidx, LX) is a function for outputting PicOrderCnt of the reference picture RefPicListX [refidx] of pic. Wherein X may be 0 or 1. The PicOrderCnt of the reference picture exists until the picture is processed as " non-exisiting ". Clip3 (a, b, c) means a function for cutting c such that it exists within a range between a and b.

ColPic with a collocated partition is RefPicList1 [0] if the slice type (slice_type) is a B-slice and collocated_from_l0_flag is 0; Otherwise, if the slice type is a P-slice or if collocated_from_l0_flag is 1, then RefPicList0 [0].

The positions of colPu and colPu (xPCol, yPCol) are derived in the following order.

1. The lower right luminance component position (xPRb, yPRb) of the current prediction unit is defined by Equation (8) and Equation (9).

2. If colPu is encoded in intraprediction mode, or colPu is not present,

(1) The center luminance component position (xPCtr, yPCtr) of the current prediction unit is defined by Equation (10) and Equation (11).

(2) colPu is set to a prediction unit that includes the positions of ((xPCtr >> 4) << 4, (yPCtr >> 4) << 4) in colPic.

3. (xPCol, yPCol) is the position of the upper left luminance component of colPu from the upper left luminance component position of colPic.

mvLXCol and availableFlagLXCol are derived as follows.

1. If colPu is encoded in intraprediction mode, or colPu is not present, then each component of mvLXCol is 0 and availableFlagLXCol is 0.

2. Otherwise, if colPu is not encoded in intraprediction mode and colPu is present, mvLXCol and refIdxCol are derived as follows:

(1) If PredFlagL0 [xPCol] [yPCol] is 0, the motion vector mvCol is determined as MvL1 [xPCol] [yPCol] and the reference picture index refIdxCol is determined as RefIdxL1 [xPCol] [yPCol].

(2) Otherwise, that is, if PredFlagL0 [xPCol] [yPCol] is 1, the following process is performed.

1) If PredFlagL1 [xPCol] [yPCol] is 0, the motion vector mvCol is determined as MvL0 [xPCol] [yPCol] and the reference picture index refIdxCol is determined as RefIdxL0 [xPCol] [yPCol].

2) Otherwise, that is, if PredFlagL1 [xPCol] [yPCol] is 1, the following procedure is performed.

a. X becomes 0 or 1, and the following allocation process is performed.

i. RefIdxColLX is assigned RefIdxLX [xPCol] [yPCol].

ii. If RefPicOrderCnt (colPic, RefIdxColLX, LX) is larger than PicOrderCnt (currPic) or PicOrderCnt (colPic) is larger than PicOrderCnt (currPic) and RefPicOrderCnt If it is smaller than PicOrderCnt (currPic), MvXCross is assigned as 1.

iii. Otherwise, if PicOrderCnt (colPic) is smaller than PicOrderCnt (currPic) and RefPicOrderCnt (colPic, RefIdxColLX, LX) is smaller than or equal to PicOrderCnt (currPic) , RefIdxColLX, LX) is greater than or equal to PicOrderCnt (currPic), then MvXCross is assigned to one.

b. When one of the following conditions is satisfied, the motion vector mvCol and the reference picture indexes refIdxCol and ListCol are determined as MvL1 [xPCol] [yPCol] and RefIdxColL1, L1, respectively.

i. Mv0Cross is 0, and Mv1Cross is 1.

ii. Mv0Cross and Mv1Cross are the same, and the reference picture list is L1.

c. Otherwise, the motion vector mvCol and the reference picture indexes refIdxCol and ListCol are determined as MvL0 [xPCol] [yPCol], RefIdxColL0, and L0, respectively.

3) availableFlagLXCol is 1, and the operations of Equation (12) or (13) to (18) are performed.

a. If PicOrderCnt (colPic) -RefPicOrderCnt (colPic, refIdxCol, ListCol) is the same as PicOrderCnt (currPic) -RefPicOrderCnt (currPic, refIdxLX, LX)

b. Otherwise,

Here, td and tb are expressed by the following equations (17) and (18).

That is, referring to equations (13) to (16), mvLXCol is derived as a scaled version of the motion vector mvCol.

On the other hand, even if the motion vector is cut into the dynamic range, if the cut motion vector is scaled, the dynamic range can be deviated again. Thus, after deriving the resized motion vector, the dynamic range of the motion vector can be limited. In this case, equations (15) and (16) can be replaced by equations (19) and (20), respectively.

IV. Information transmission method for cutting a temporal motion vector in a decoding apparatus

Hereinafter, a method of transmitting information necessary for cutting a temporal motion vector in the decoding apparatus in the same manner as the encoding apparatus will be described.

The TMVBitWidth in the above-described temporal motion vector derivation process can be transmitted from the encoding apparatus to the decoding apparatus through a sequence parameter set, a picture parameter set, or a slice header.

Bit_width_temporal_motion_vector_minus8 in Table 5 indicates the bit width of the temporal motion vector component. If bit_width_temporal_motion_vector_minus8 does not exist, it is inferred as 0, and the bit width of the temporal motion vector component is expressed as shown in Equation (21).

1. Information transmission method 1 - When the motion vector is compressed and the bit depth of the motion vector is limited

Referring to Table 6, when motion_vector_buffer_comp_flag is 1, a motion vector buffer compression process is performed.

motion_vector_buffer_comp_ratio_log2 represents the compression ratio of the motion vector buffer compression process. If motion_vector_buffer_comp_ratio_log2 does not exist, it is inferred as 0, and the motion vector buffer compression ratio is expressed by Equation (22).

Referring back to Table 6, when bit_depth_temporal_motion_vector_constraint_flag is 1, a temporal motion vector bit depth limitation process is performed.

bit_depth_temporal_motion_vector_minus8 indicates the bit depth of the temporal motion vector. If bit_depth_temporal_motion_vector_minus8 does not exist, it is inferred as 0, and the bit depth of the temporal motion vector is expressed by Equation (23).

2. Information transmission method 2 - Limiting bit depth of motion vector

Referring to Table 7, when the bit_depth_temporal_motion_vector_constraint_flag is 1, a temporal motion vector bit depth limitation process is performed.

bit_depth_temporal_motion_vector_minus8 indicates the bit depth of the temporal motion vector. If bit_depth_temporal_motion_vector_minus8 does not exist, it is inferred as 0, and the bit depth of the temporal motion vector is expressed by Equation (24).

3. Information transmission method 3 - Limiting bit depth of motion vector

Bit_depth_temporal_motion_vector_minus8 in Table 8 indicates the bit depth of the temporal motion vector. If bit_depth_temporal_motion_vector_minus8 is not present, it is inferred as 0, and the bit depth of the temporal motion vector is expressed as:

4. Information transmission method 4 - When the bit depth is limited for each of the X component and Y component of the motion vector

Referring to Table 9, when the bit_depth_temporal_motion_vector_constraint_flag is 1, the temporal motion vector bit depth limitation process is performed.

bit_depth_temporal_motion_vector_x_minus8 represents the X component bit depth of the temporal motion vector. If bit_depth_temporal_motion_vector_x_minus8 does not exist, it is inferred as 0, and the bit depth of the temporal motion vector is expressed by Equation (26).

bit_depth_temporal_motion_vector_y_minus8 represents the Y component bit depth of the temporal motion vector. If bit_depth_temporal_motion_vector_x_minus8 does not exist, it is deduced to be 0, and the bit depth of the temporal motion vector is expressed by Equation (27).

5. Information transmission method 5 - When the motion vector is compressed and the bit depth of the motion vector is limited

Referring to Table 10, when motion_vector_buffer_comp_flag is 1, a motion vector buffer compression process is performed.

motion_vector_buffer_comp_ratio_log2 represents the compression ratio of the motion vector buffer compression process. If motion_vector_buffer_comp_ratio_log2 does not exist, it is deduced to be 0, and the motion vector buffer compression ratio is expressed by Equation (28).

V. Definition of dynamic range through video codec level

The dynamic range of a temporal motion vector is not transmitted through a sequence parameter set, a picture parameter set, or a slice header, but may be defined through a level of an image codec. The encoding apparatus and the decoding apparatus can discriminate the limited dynamic range of the motion vector using the level information.

Further, the dynamic range and / or bit depth of each of the X component and the Y component of the motion vector may be defined differently from the level, and the minimum value and the maximum value of each component may be defined.

Tables 11 and 12 are examples of a case where the TMVBitWidth in the process of deriving the temporal motion vector is defined at a level.

Referring to Table 11, TMVBitWidth is set to MaxTMVBitWidth defined in the level. At this time, MaxTMVBitWidth represents the maximum bit width of the motion vector when the temporal motion vector is stored in the memory.

On the other hand, TMVBitWidth may be defined at a level, and a sequence parameter set, a picture parameter set, or a slice header may be transmitted as a delta value from a defined value. That is, TMVBitWidth may be set to MaxTMVBitWidth defined in the level plus the difference set in the sequence parameter set, the picture parameter set, or the slice header. At this time, TMVBitWidth represents the bit width of the motion vector when the temporal motion vector is stored in the memory.

Delta_bit_width_temporal_motion_vector_minus8 in Table 13 indicates the difference in bit width of temporal motion vector components. If delta_bit_width_temporal_motion_vector_minus8 does not exist, it is deduced to be 0, and the bit width of the temporal motion vector component is expressed by Equation (29).

It is also possible to define the dynamic range of each component of the temporal motion vector in the level as shown in Table 14. [

It is also possible to define the bit width of each component of the temporal motion vector in the level as shown in Tables 15 to 17. [

It is also possible to define the bit width of the temporal motion vector Y component in the level as shown in Table 18.

In addition, the dynamic range of the temporal motion vector may be defined as a fixed value predetermined by the encoding apparatus and the decoding apparatus, or may be stored in the form of a fixed bit depth, without transmitting the information on the motion vector limitation.

TMVBitWidth may be a positive integer such as 4, 6, 8, 10, 12, 14, 16, etc. when TMVBitWidth is fixed to the same value in the encoding apparatus and the decoding apparatus. At this time, TMVBitWidth represents the bit width of the motion vector when the temporal motion vector is stored in the memory.

14 is a flowchart illustrating a method of encoding an image according to an embodiment of the present invention. Referring to FIG. 14, the image encoding method includes a cutting step S1410, a storing step S1420, and an encoding step S1430.

The image coding apparatus and / or the decoding apparatus cuts the motion vector of the reference picture into a predetermined dynamic range (S1410). As described above through "I. Motion Vector Cutting Process ", a motion vector out of the dynamic range is expressed by a minimum value or a maximum value of the corresponding dynamic range. Therefore, as described above, the level and / or the sequence parameter of the video codec, as described above, through " Information transmission method for cutting temporal motion vector in IV. Decoding apparatus "and" V. Definition of dynamic range through level of video codec & A motion vector of a reference picture can be cut to a predetermined dynamic range by restricting the bit depth through a set or the like or restricting the dynamic range through the level of the video codec.

The video encoding apparatus and / or the decoding apparatus stores the motion vector of the cut reference picture in the buffer through the " II. Motion vector storing process "(S1420). The motion vector may be stored in the buffer together with the reconstructed image or separately.

The image encoding apparatus encodes the motion vector of the current block using the motion vector of the stored reference picture (S1430). As described above with reference to "III. Motion vector derivation process ", in the enhanced motion vector prediction method used in the HEVC, not only the motion vector of the restored block located in the periphery of the current block to be coded / decoded, A motion vector of a block existing at the same position or corresponding position with the target block is used. Accordingly, the motion vector of the current block may be a motion vector of a reference picture, that is, a temporal motion vector, as well as a motion vector of a neighboring block adjacent to the current block.

On the other hand, since the dynamic range of the X component and the dynamic range of the Y component of the motion vector of the reference picture can be defined differently, each component of the motion vector of the reference picture can be cut into each dynamic range.

It is also possible to use a method of compressing a motion vector of a reference picture as well as a method of restricting the dynamic range of a motion vector of a reference picture. When limiting the dynamic range of a motion vector of a reference picture or compressing a motion vector of a reference picture, a flag indicating the level and / or a sequence parameter set of the video codec and parameters related thereto may be defined.

In addition, it is also possible to perform a coding method such as motion vector prediction, enhanced motion vector prediction, motion information merging, and motion information merge skip using motion information stored in the memory, i.e., motion information of a reference picture.

15 is a flowchart illustrating a method of decoding an image according to an embodiment of the present invention. Referring to FIG. 15, the image decoding method includes a cutting step S1510, a storing step S1520, a deriving step S1530, and a decoding step S1540.

The cutting step S1510 and the storing step S1520 of FIG. 15 are the same as the cutting step S1410 and the storing step S1420 of FIG. 14 using the " I. Motion vector cutting process " . The derivation step S1530 of FIG. 15 uses the above-described " III. Motion vector derivation process "and is symmetric with the encoding step S1430 of FIG. Therefore, a detailed description will be omitted.

The video decoding apparatus performs inter prediction decoding using the motion vector of the current block to be decoded (S1540). The video decoding apparatus stores a motion vector in a memory using at least one of a dynamic range limitation method of a motion vector, a spatial resolution reduction method of a motion vector, a motion vector quantization method, and a motion vector resolution reduction method, The vector can be used for motion vector prediction and motion information merging of the decoding target block.

In addition, it is also possible to perform decoding methods such as motion vector prediction, enhanced motion vector prediction, motion information merging, and motion information merge skip using motion information stored in the memory, that is, motion information of a reference picture.

Although the above-described embodiments are described through a series of steps or a flowchart represented by blocks, the present invention is not limited to the order of the steps described above, and some steps may occur in different steps, different orders, or concurrently. It will also be understood by those skilled in the art that the steps shown in the flowchart illustrations are not exclusive, that other steps may be included, or that some steps may be deleted.

In addition, the above-described embodiments include examples of various aspects. While not all possible combinations can be described to illustrate various aspects, one of ordinary skill in the art will recognize that other combinations are possible. Accordingly, it is intended that the invention include all alternatives, modifications and variations that fall within the scope of the following claims.

Claims (20)

A reference picture buffer for storing a reference picture; And
And a motion compensation unit for generating a prediction block using a motion vector of the reference picture and the reference picture,
The motion vector of the reference picture is cut to a predetermined range,
Wherein the motion vector is cut into a predetermined fixed value range.
The method according to claim 1,
Wherein the motion vector of the reference picture is scaled and then cut to the range.
delete The method according to claim 1,
The motion vector of the reference picture is stored in units of a predetermined block,
Wherein the motion compensation unit generates the prediction block using a motion vector of the reference picture stored in the predetermined block unit.
The method according to claim 1,
Wherein the x component and the y component of the motion vector are cut to have the same fixed value range.
The method according to claim 1,
Wherein the motion vector is a motion vector of a block decoded in the inter prediction mode. delete delete delete delete delete delete delete delete delete delete delete delete delete delete

Images